The country must also develop MOX fuel fabrication and fast breeder reactors to achieve its ambitious goals
By Dan Yurman, Contributing Reporter
Fuel Cycle Week (V10:N406) January 6, 2011
China announced Jan. 3 it has begun to reprocess spent nuclear fuel. The technology, tested at a plant operated by state-owned China National Nuclear Corp. (CNNC) at a remote site in the Gobi desert in Gansu province, enables the re-use of materials from irradiated fuel from light water reactors.
Wang Junfeng, CNNC project director, said the process is "effective and safe." According to China Central Television (CCTV), Junfeng said the country now had enough fuel to last up to 70 years and the new technology could yield enough to last for 3,000 years.
This statement on its face seems to border on fantasy, given what is known about China's investment in other parts of the fuel cycle, including MOX fuel fabrication and development of fast breeder reactors.
It is a fact that China has been stockpiling uranium for its hugely ambitious, 80 GWe nuclear reactor construction plan over the next decade. The country’s recent multi-million pound purchases have contributed to a steady rise in the U3O8 spot price, which stands at $62.50/pound a $22.50/pound increase from this time last year.
There are a few confirmed details about China’s reprocessing technology so far, although CCTV made a point of saying the process is a secret.
In 2006, for example, a pilot reprocessing plant (50 tonnes/year) using the Purex process was opened in 2006 at Lanzhou Nuclear Fuel Complex. It is reported to be capable of processing 100 tonne/year.
Part of AREVA Deal
Given that in 2007 CNNC inked an agreement with AREVA to begin a feasibility study to build a spent fuel reprocessing plant, Fuel Cycle Week believes this week’s announcement is not so much a breakthrough as the start of development of a commercial-sized reprocessing plant of 800- to-1,000-tonne/year that uses French technology.
Last November the two companies signed an industrial agreement for the design, construction and commissioning of the plant, the World Nuclear Association reports. The plant could be sited at Lanzhou or Jiayuguan, both in Gansu province.
An 800 tonne/year spent fuel reprocessing plant is thought by some experts to be about as large a plant as can be efficiently managed with current technology. The period needed to construct such a plant could be at least a decade. It would take a few years to plan and at least five more to build, test, and reliably process fuel in volume.
It is unlikely that a first of a kind facility would start up without problems. For instance, Japan Nuclear Fuels has experienced repeated delays at Rokkasho, with the latest occurring in October 2010 setting back the start-up date by at least two years.
Once the fuel is reprocessed, China still needs a new fuel fabrication facility to make the MOX fuel assemblies and a permanent geologic repository for the remaining highly radioactive waste products that can't be recycled. None of these types of facilities are named in China's announcement this week. As with fuel reprocessing, the time line to reliable production could easily consume a decade.
Last October Reuters reported that Tractabel, the Belgian subsidiary of French utility GDF Suez, together with Belgonucleaire and the research center SCK-CEN, signed a framework agreement with CNNC to build a pilot installation for MOX fuel manufacturing.
There was no word at the time whether the agreement would conflict with AREVA’s contractual relationships in China. Its 2007 agreement with CNNC also included a feasibility study for a MOX fuel fabrication plant.
According to CNNC official Sun Donghui, separated plutonium would initially be used in MOX fuel for an experimental 25 MWe fast breeder reactor. The MOX fuel could be burned in conventional LWRs which would serve to produce a modest, but certainly not a 3,000-year extension, in China's fuel supply.
Big Bet on Fast Reactors
The Chinese statement about expectations for an enormous extension of uranium supplies strongly suggests fast breeder reactors would have to be in the picture, as MOX fuel does not completely close the fuel cycle.
Fast breeder reactors, which produce fuel, use liquid metal-cooled systems such as sodium or a lead-bismuth combination. Both reactor types are reference designs in the international Gen IV R&D consortium,
However, scientists at the Department of Energy's Idaho National Laboratory told Fuel Cycle Week in Summer 2009 that commercial deployment of these types of fast reactors is easily two decades or more in the future.
Even if China has, as it claims, 70 years of fuel, can it develop a reliable fleet of 1,000 MW fast breeder reactors and associated fuel facilities in 20 years? Can it sustain the political will and funding to do so over such a long period of time?
What we know and don’t know
Here's what we do know about China's work on fast reactors. Last summer the Denki Shimbun, an English language Japanese news service, reported that China plans to drive the development of its fast reactor program to produce new units in the power range of 1,000-1,500 MW by 2020. This date would seem to be unrealistic given the status and timeline for development of Gen IV designs in the West.
However, last July China achieved criticality at its first prototype fast neutron reactor. The Chinese Experimental Fast Reactor is expected to reach a thermal capacity of 60 MW and produce 20 MW of electric power. It is a long way from a 20 MW prototype to a 1000 MW commercial version.
Developed by the China Institute of Atomic Energy, it is the first sodium-cooled fast reactor in the country. The reactor was built in collaboration with several Russian entities including OKBM Afrikantov, OKB Gidropress, NIKIET, and the Kurchatov Institute.
It appears that China has set aside plans for a 600 MW indigenous design in favor of buying two BN-800 fast reactors from Russia for Sanming-1 and -2. The project is expected to break ground in August 2011.
A bilateral program on fuel cycles for fast reactors is part of the effort. China will build a nuclear city around the two reactors to house construction workers, reactor operators, and support services.
Nuclear industry experts tell FCW the fast reactor deal between the two countries is not yet a done deal. Perhaps reflecting this development, Bloomberg cited a statement last July from Martin Wang, an energy analyst at Guotai Junan, based in Hong Kong, who said it will take China "some time" to develop its fast reactors for commercial use.
That view was countered by one from Steve Kidd, head of strategy at the World Nuclear Association, who told Bloomberg in July 2010 the technology could be in commercial use by 2025.
"If China is doing 10 PWRs a year, there is a big economic inventive to do something better. The technology could come earlier than people think."
Fact Box: China's uranium needs in 2020
In 2010 China's uranium consumption for its 9 GWe of nuclear reactor generating capacity was just under 2,000 tonnes. China has published an estimate that by 2020 it will have built out to 80 GWe or nearly nine times the current installed based. Assuming all the reactors are light water designs, a straight linear projection would indicate China would have demand for 20,000 tonnes of uranium a year.
The caveat on the estimate is that it would be affected by how much MOX manufacturing China can produce by 2020. For instance, what would the yield be in materials suitable for new fuel from a reprocessing plant that could handle 800 tonnes/year of spent fuel? The key to China's reported "breakthrough" would be proven, or not, by this number.
According to published data on French spent fuel reprocessing and MOX manufacturing, 800 tonnes of spent fuel would produce about 8 tonnes of plutonium and 760 tonnes of uranium. The two would be combined with enriched uranium to produce MOX fuel equal to uranium fuel at 4.5% u235. These are round numbers and actual process outputs will vary according to customer requirements.
# # #